ORIGINAL PAPER

Characterization, efficient transformation and regenerationof Chirita pumila (Gesneriaceae), a potential evo-devo model plant

Bo-Ling Liu • Xia Yang • Jing Liu •

Yang Dong • Yin-Zheng Wang

Received: 12 December 2013 / Accepted: 4 April 2014

� Springer Science+Business Media Dordrecht 2014

Abstract An efficient transformation and regeneration

system is essential for functional investigation of devel-

opmental genes and related elements in the field of evo-

lutionary developmental biology (evo-devo). Chirita

pumila D. Don belongs to the Gesneriaceae family, one of

the most basal groups in Lamiales sensu lato, and possesses

many tractable biological features including annual habit,

small plant size, short generation time, abundant offspring

and low chromosome number. In addition, C. pumila has

cleistogamous flowers with potential cross-pollination, a

special phenomenon first reported herein in Gesneriaceae.

Parameters affecting shoot induction and genetic transfor-

mation have been evaluated, including plant growth regu-

lators, temperature, antibiotic concentration, pre- and co-

culture duration, Agrobacterium cell density and infection

time. Polymerase chain reaction and b-Glucuronidase

(GUS) activity assays of T0 and T1 plants show that the

GUS gene has been introduced into the host with stable and

universal expression. The applicability of the transforma-

tion system in gene function investigation is further con-

firmed by transforming a GsNST1B gene from Glycine

soja. This transformation system provides a valuable

platform for deep function analyses of related genes and

elements for a wide range of evo-devo studies, especially

in the field of floral evolution, which would develop its

potential of being a model organism in Lamiales s. l.

Keywords Chirita pumila � Cleistogamy � Evo-devo �Genetic transformation � Gesneriaceae

Abbreviations

BA 6-Benzylaminopurine

GFP Green fluorescent protein

GUS b-Glucuronidase

HPTII Hygromycin phosphotransferase gene

MS Murashige and Skoog medium

NAA 1-Naphthaleneacetic acid

PCR Polymerase chain reaction

RT-PCR Reverse transcription-PCR

Introduction

Only after the rise of evolutionary developmental biology

(evo-devo) has the integration of developmental processes

and genetic and evolutionary biology at the molecular level

allowed the analysis of how developmental processes can

result in morphological evolution (Breuker et al. 2006;

Kellogg 2006; Muller 2007; Carroll 2008; Kopp 2009; de

Bruijn et al. 2012). Over the past two decades, evo-devo as

an emerging biological discipline has made considerable

achievements in discovering extensive similarities in gene

regulation among distantly related species with dramati-

cally different body plans in both animals and plants

relying on rapid technical advancements in gene clone and

Bo-Ling Liu and Xia Yang have contributed equally to this work.

Electronic supplementary material The online version of thisarticle (doi:10.1007/s11240-014-0488-2) contains supplementarymaterial, which is available to authorized users.

B.-L. Liu � X. Yang � J. Liu � Y. Dong � Y.-Z. Wang (&)

State Key Laboratory of Systematic and Evolutionary Botany,

Institute of Botany, Chinese Academy of Sciences,

20 Nanxincun, Xiangshan, Beijing 100093, China

e-mail: wangyz@ibcas.ac.cn

B.-L. Liu

Qufu Normal University, Qufu, Shandong, China

123

Plant Cell Tiss Organ Cult

DOI 10.1007/s11240-014-0488-2

expression. Examples include Hox genes in establishing the

anterior–posterior axis in bilaterian animals, MADS-box

genes in patterning floral organ identities and CYCLOIDEA

(CYC)-like TCP genes in determining floral zygomorphy

(Breuker et al. 2006; Kellogg 2006; Muller 2007; Carroll

2008; Kopp 2009; de Bruijn et al. 2012). These studies

have provided remarkable insights into the evolutionary

conservation of developmental programs and the mecha-

nisms underlying modification of developmental processes

that generate morphological novelties (Stern 2000; Pruitt

et al. 2003; Breuker et al. 2006). Currently, most evo-devo

studies in plants, especially outside model organisms,

depend on global DNA sequence analyses and correlative

analyses of candidate gene expression to corresponding

morphologies rather than gene function investigation. Even

though gene expression studies are sufficient ways to

screen genes for putative regulatory changes, they are

association rather than causality analyses (Baguna and

Garcia-Fernandez 2003; Kellogg 2004). Therefore, it is

essential and critically important to conduct comparative

functional studies to demonstrate that such regulatory

changes are actually responsible for phenotypic variations

and to gain an integrated view of the role of development

in evolution (Irish and Benfey 2004; Breuker et al. 2006).

As functional analyses become widespread in evo-devo

studies, researchers usually transfer target genes into a

distantly related classical model organism to test their

function because of the difficulty in carrying out such

experiments in native systems (Irish and Benfey 2004).

However, these gene transfers are not always efficient to

test the genes’ function or may not reflect their actual

function in native contexts (Irish and Benfey 2004).

Therefore, evo-devo biologists have increasingly recog-

nized the limitation of the classical model organisms and

the urgency to develop new model organisms to efficiently

investigate the genes’ actual function in specific morpho-

logical novelties (Mandoli and Olmstead 2000; Irish and

Benfey 2004; Jeffery 2008).

In angiosperms, one key innovation is the occurrence of

the flower with subsequent remarkable diversification upon

wide modifications of the genetic programs controlling

floral organ identity, floral symmetry and reproduction

system (Dilcher 2000; Kramer 2007). Currently, the focus

of plant evo-devo studies is mainly on the evolution and

diversity of ABCE model beneath floral organ identity and

gene network underlying floral symmetry first identified

and elaborately studied in classical model species Arabi-

dopsis and Antirrhinum (Irish and Benfey 2004; Kramer

2007). New evo-devo model organisms would yield new

insights into the origin of major floral evolutionary nov-

elties in particular lineage histories that could not be tar-

geted by classical model organisms. The Gesneriaceae

family is one of the most basal groups in Lamiales sensu

lato (Endress 1998; Cubas 2004; Wortley et al. 2005; http://

www.mobot.org/MOBOT/Research/APweb/welcome.html),

a major angiosperm clade predominant with zygomorphic

flowers that are believed ancestral in this order (Donoghue

et al. 1998; Cubas 2004). Therefore, Gesneriaceae locates at

a phylogenetic node of floral evolution in angiosperms. As a

member of Gesneriaceae, Chirita pumila D. Don is a

promising candidate of model species for evo-devo studies

in floral evolution because it shares a series of biological

features with classical model plants, such as annual habit,

diploid with the lowest chromosome number (2n = 8) in

Gesneriaceae (Ratter 1963; Li and Wang 2004; this study),

and cleistogamy with potential cross-fertilization (see

results in this study). In addition, the whole genome

sequencing project is carrying out (Yi-Kun He, personal

communication). These unique biological features give

C. pumila a great advantage in tractability for laboratory

experiments over other Gesneriaceae species that are usually

perennial and polyploidy with cross-pollination, including

the two famous ornamental plants Saintpaulia ionantha and

Sinningia speciosa and the physiological model plant Ra-

monda myconi successful in Agrobacterium-mediated

genetic transformation (Mercuri et al. 2000; Kushikawa

et al. 2001; Toth et al. 2006; Zhang et al. 2008).

Our laboratory has conducted a series of evo-devo

studies relating to the evolution of floral symmetry in

Gesneriaceae (Du and Wang 2008; Gao et al. 2008; Zhou

et al. 2008; Song et al. 2009; Pang et al. 2010; Yang et al.

2010; Liu et al. 2014) and the molecular mechanism

underlying the repeated origins of floral zygomorphy in

angiosperms (Yang et al. 2012). Recently, deep functional

analyses of related gene networks in floral symmetry and

floral organ identity are carrying out in C. pumila and its

relatives (our unpublished results). Herein, we report an

efficient Agrobacterium-mediated transformation and

regeneration system developed in C. pumila by evaluating

several factors affecting shoot induction and genetic

transformation and validating its applicability in gene

function investigation using the GsNST1B gene functioning

in secondary wall biosynthesis in Glycine soja. This

transformation system would have wide applications in the

field of evo-devo studies.

Materials and methods

Plant material and culture conditions

The C. pumila plants, collected from Hekou County,

Yunnan, China (Wang, HK01), were grown in 8 cm pots

containing the mixture of vermiculite and Pindstrup

substrate (Pindstrup) (1:2) in culture room. The growth

conditions were: 26 �C, a 10/14 light/dark photoperiod

Plant Cell Tiss Organ Cult

123

under cool-white fluorescent light of 100 lmol m-2 s-1

and 50–70 % of relative humidity. Voucher specimens

were deposited in the Herbarium, Institute of Botany,

Chinese Academy of Sciences.

Agrobacterium strain and plasmids

Agrobacterium tumefaciens strain LBA4404 harboring the

binary vector pCAMBIA1301 was used in parameter eval-

uation experiments and b-Glucuronidase (GUS) activity

assay. The vector carries the hygromycin phosphotransfer-

ase (HPTII) gene for transformant selection on hygromycin

and the GUS reporter gene that is interrupted by an intron

(Fig. 1a). The p35S::GsNST1B plasmid was constructed as

described (Dong et al. 2013). Briefly, the full-length coding

sequence of GsNST1B gene (or GsSHAT1-5; Dong et al.

2013, 2014) was amplified (50-GGAAGATCTGCCGGA

AAACATGAG-30 and 50-GGACTAGTCTACACTG ACG

TGTTGGAC-30), digested with BglII and SpeI, and inserted

into the binary vector pCAMBIA1302 that contains the

HPTII gene and the green fluorescent protein (GFP) gene

(Fig. 1b). The resultant construct was introduced into

Agrobacterium LBA4404 by electroporation (Eppendorf).

Pollen germination assay and aniline blue staining

of pollen tubes

Pollen germination experiment was performed according to

Mori et al. (2006). Briefly, pollen was randomly collected

from six different flowers close to anthesis and dispersed into

sterilized water. 15 ll of the suspension was carefully

flattened onto the culture medium (containing 150 g l-1

sucrose, 40 mg l-1 boric acid, 20 mg l-1 calcium chloride,

6 g l-1 agar), cultured at 28 �C in the dark for 2–8 h and

examined using a Zeiss Axio Imager A1 M Microscope

(Zeiss).

To perform aniline blue staining, flowers close to

anthesis were either collected directly or bagged in Cel-

lophane for further 2 days. The pistils were dissected and

stained with aniline blue according to Jiang et al. (2005).

Briefly, the pistils were fixed in ethanol: chloroform: acetic

acid (6:3:1) for 24 h, softened in 8 M NaOH for 6 h and

washed three times with 0.1 M K2HPO4-KOH buffer (pH

7.5). Then, the pistils were stained in 0.1 % aniline blue

solution (pH 11) in the dark for 4 h and observed with a

Leica TCS SP5 Fluorescence Microscope (Leica).

Karyotype analysis and measurement of the genome

size

Root tips were pretreated with a mixture of 2 mM

8-hydroxyquinoline and 0.1 % colchicine (1:1) at 20 �C for

4 h, and fixed in Carnoy’s I (100 % ethanol and glacial

acetic acid, 3:1) at 5 �C for 1 h. The fixed materials were

macerated in 1 M HCl at 60 �C for 1 min, stained with

carbol fuchsin, squashed and photographed under a

microscope. The length of long and short arms of each

metaphase chromosome was measured, and the chromo-

some arm ratio was estimated by the length of long arm/the

length of short arm. The relative length was calculated by

the length of individual chromosome/the length of all

chromosomes 3100 %.

Fig. 1 Map of the binary vector

pCAMBIA1301 and the

p35S::GsNST1B construct.

a T-DNA region of the

pCAMBIA1301 vector. b The

binary vector pCAMBIA1302

and the constructed

p35S::GsNST1B plasmid

Plant Cell Tiss Organ Cult

123

Flow cytometry was used to measure the genome size of

C. pumila according to Dolezel et al. (2007). Rice (Oryza

sativa L. var. Nipponbare) was used as an internal standard.

Briefly, young leaves of the sample and reference standard

were chopped quickly with a sharp razor blade in a plastic

Petri dish containing 1 ml ice-cold Galbraith’s buffer

(45 mM MgCl2, 20 mM MOPS, 30 mM sodium citrate,

0.1 % Triton X-100, pH 7.0) (Galbraith et al. 1983). The

resultant homogenate was filtered through a 25 lm nylon

mesh to remove large debris, and incubated in staining

solution containing 50 g l-1 propidium iodide (Sigma-

Aldrich) and 50 g l-1 RNaseA (TaKaRa) on ice in the dark

for 20 min. The relative nuclear DNA fluorescence inten-

sity was measured using a MoFlo� High-performance Cell

Sorter (Beckman). Three C. pumila plants were analyzed

with three replicates each.

Seed germination, shoot induction and antibiotic

sensitivity experiments

Chirita pumila plants were grown to flowering stage and

seeds were harvested. The seeds were surface-sterilized in

70 % ethanol for 1 min, rinsed with sterile water once,

disinfected with 2.5 % sodium hypochlorite for 3–5 min

and finally rinsed five times with sterile water. The steril-

ized seeds were germinated on Murashige and Skoog (MS)

medium (Murashige and Skoog 1962) in a growth chamber

at 26 �C under a photoperiod of 10/14 h light/dark

(100 lmol m-2 s-1). About 2-month-old plantlets were

used for preparing leaf explants.

To determine optimal concentration of growth regula-

tors for shoot induction, fresh leaf explants were cultured

on MS medium containing different concentrations of

6-benzylaminopurine (BA) and a-naphthalene acetic acid

(NAA) (see Table 1) at 26 �C. To evaluate the effect of

temperature on shoot induction, fresh leaf explants were

cultured on MS medium containing 0.5 mg l-1 BA and

0.1 mg l-1 NAA (based on the result of growth regulator

experiment; see Table 1) at 22, 24, 26 or 28 �C (see Sup-

plementary Table 4). To evaluate whether hygromycin is

appropriate for selecting transformants, fresh leaf explants

were incubated on MS medium containing 0.5 mg l-1 BA,

0.1 mg l-1 NAA and different concentrations of hygro-

mycin (0, 5, 10, 15, 20 and 30 mg l-1; see Supplementary

Table 5) at 26 �C. For each experiment, the explants were

always maintained on the same medium without sub-

culture, and the shoot induction rate was calculated

4 weeks later. Data, presented as mean ± SD, were cal-

culated from three independent experiments with about 40

leaf explants each. Means of induction efficiencies were

compared for level of significance (P \ 0.05) using a

Fisher’s Least Significant Difference (LSD) test.

Evaluation of parameters affecting shoot induction rate

in transformation experiments

Four parameters were successively evaluated, including co-

culture time, pre-culture duration, Agrobacterium cell

density and infection time. In each experiment, only one

factor was changed with other fixed. The following is a

general protocol for these experiments. Fresh leaf explants

were pre-cultured on MS medium containing 0.5 mg l-1

BA and 0.1 mg l-1 NAA for 0, 1, 2, 3 or 4 days. Agro-

bacterium LBA4404 cells (harboring pCAMBIA1301)

cultured in YEB medium (containing 100 mg l-1 strepto-

mycin, 50 mg l-1 rifampicin and 50 mg l-1 kanamycin) at

28 �C overnight were inoculated to fresh YEB medium and

grown to OD600 = 0.2, 0.4, 0.6, 0.8 or 1.0. The cells were

harvested by centrifugation at 5,000 rpm for 10 min, rinsed

with MS liquid once, and resuspended in MS liquid con-

taining 150 mg l-1 acetosyringone. The harvested cells

were used to inoculate pre-cultured explants for 10, 20, 30,

40 or 50 min. After being briefly blot-dried with sterile

filter papers, the explants were incubated on the co-culture

medium containing 0.5 mg l-1 BA, 0.1 mg l-1 NAA and

150 mg l-1 acetosyringone at 26 �C in the dark for 1, 2, 3,

4 or 5 days, and then transferred to the shoot induction

medium containing 0.5 mg l-1 BA, 0.1 mg l-1 NAA,

Table 1 Effects of different concentrations of BA and NAA on shoot

induction rate of C. pumila leaf explants

BA

(mg l-1)

NAA

(mg l-1)

Shoot induction

rate (%)

Root induction

rate (%)

0 0 67.8 ± 10.0 0

1 1 72.7 ± 7.3 42.5 ± 11.5

1 0.5 79.9 ± 5.1 15.0 ± 0.6

1 0.2 84.8 ± 8.0 0

1 0.1 87.2 ± 11.8 0

0.5 1 87.6 ± 4.2 92.5 ± 0.3

0.5 0.5 82.2 ± 9.2 42.3 ± 6.7

0.5 0.2 92.5 ± 0.3 0

0.5 0.1 97.6 – 4.1 0

0.2 1 65.4 ± 13.9 87.4 ± 4.8

0.2 0.5 60.1 ± 8.1 32.2 ± 10.0

0.2 0.2 85.0 ± 0.6 5.0 ± 4.3

0.2 0.1 82.4 ± 4.8 0

0.1 1 89.7 ± 11.8 95.1 ± 4.3

0.1 0.5 59.9 ± 12.2 35.0 ± 3.9

0.1 0.2 89.9 ± 4.6 7.5 ± 0.3

0.1 0.1 74.9 ± 5.0 0

Data are presented as mean ± SD collected from three independent

experiments, including a total of 40 leaf explants per replicate

The optimal concentration of BA and NAA for the shoot induction is

highlighted by bold letters

Plant Cell Tiss Organ Cult

123

20 mg l-1 hygromycin and 300 mg l-1 carbenicillin. The

hygromycin-resistant shoot induction rate was evaluated

4 weeks later. Data (mean ± SD) were calculated from

three independent experiments with about 40 leaf explants

each. Means of induction efficiencies were compared for

level of significance (P \ 0.05) using a Fisher’s LSD test.

GUS activity assay

One-day pre-cultured leaf explants were inoculated with

Agrobacterium LBA4404 (harboring pCAMBIA1301) for

20 min, cultured on the co-culture medium in the dark for

2 days and then transferred to the shoot induction medium.

Hygromycin-resistant shoots of about 0.5 cm in length were

excised, transferred to fresh MS medium (without growth

regulator and antibiotic), and finally transferred to pots and

grown in culture room (the growth conditions were the same

as described above). Genomic DNA was isolated from the

leaves of putative T0 plants using the Rapid Plant DNA

Extraction Kit (Tiangen), and PCR was conducted to in-

dentify positive transgenic plants using primers spanning the

35S promoter and the GUS gene (50-GTGAGCGGATAACA

ATTTCAC-30 and 50-CGAGTCGTC GGTTCTGTAAC-30).PCR conditions were: 94 �C 3 min, 30 cycles of 94 �C 30 s,

60 �C 30 s and 72 �C 60 s, and 72 �C 10 min. Plasmid and

wild-type plants were used as positive and negative controls,

respectively. GUS staining was conducted as described

(Jefferson et al. 1987). Briefly, leaves and stems of three

independent T0 transgenic plants were incubated in GUS

staining solution (50 mM sodium phosphate buffer,

0.05 mM potassium ferricyanide, 0.05 % Triton X-100,

2 mM X-Gluc, pH 7.0) at 37 �C overnight and observed

under a microscope. Wild-type plants were served as nega-

tive controls to exclude the possibility of endogenous GUS

expression. To validate whether the transformed GUS gene

could be inherited, T1 progenies of one T0 plant were dis-

infected and cultured on MS selection medium containing

25 mg l-1 of hygromycin. About 4 weeks later, the segre-

gation ratio was calculated by counting the number of ger-

minated and well developed seedlings and the number of

germinated but withered seedlings. The GUS activity assays

of T1 progenies were as described above.

Expression and histochemical analyses of GsNST1B

gene in transgenic plants

Agrobacterium LBA4404 harboring the p35S::GsNST1B

plasmid was used to infect C. pumila leaf explants. Positive

transformants were confirmed by PCR followed by DNA

sequencing to exclude the possible amplification of

endogenous NST1B-like genes. RT-PCR was conducted to

measure GsNST1B expression in transgenic leaves using

specific primers (50-CTGGCCGCGACAAAGTCATC-30

and 50-CTTCTTCCTGAGCAGCATCCG-30; Dong et al.

2013) under the following conditions: 94 �C 3 min, 30

cycles of 94 �C 30 s, 56 �C 30 s and 72 �C 30 s, and 72 �C

10 min. RT-PCR products were sequenced. As a reference

gene, CpACTIN was amplified with 26 cycles using specific

primers (50-AGTTATCACCATTGCC GCCGAGAGG-30

and 50-GCAATGCCAGGGAACATAGTCGACC-30). RT-

PCR products were visualized on a 1.5 % agarose gel.

The ectopic deposition of lignin was examined as described

(Dong et al. 2013). Briefly, transgenic leaves were fixed,

embedded in Paraplast Plus (Sigma-Aldrich) and stained

with 0.2 % toluidine blue solution. The autofluorescence of

secondary cell walls was detected using a fluorescence

microscope (Zeiss). Three transgenic plants were examined

with wild-type ones served as negative controls.

Results

Biological characteristics of C. pumila plants

Chirita pumila D. Don, an annual herb with erect stems of

6–46 cm in height, extensively distributes in Southwest

China, North India, Vietnam, Nepal, Sikkim, Bhutan, Myan-

mar and Thailand (Wang et al. 1998; Li and Wang 2004). Its

typical characteristics include purple-spotted oval leaves and

large purplish zygomorphic flowers (Fig. 2a). The capsule is

6–12 cm in length (Fig. 2a, b) that can yield abundant tiny

Fig. 2 Morphology of C. pumila. a A C. pumila plant with typical

zygomorphic flowers. Bar, 2 cm. b The capsules of C. pumila. Bar,

2 cm. c The capsule and seeds of C. pumila. Bar, 1 cm. d The

enlarged view of c. Bar, 0.1 cm

Plant Cell Tiss Organ Cult

123

spindly seeds (over 1,000 seeds per capsule based on a rough

estimate) (Fig. 2c, d). In addition, C. pumila has a short gen-

eration time (about 5 months from seed to seed).

Self-compatibility is critical to a genetic transformation

system (Bliss et al. 2013). We have noticed that C. pumila

flowers always autonomously bear fruits without any source

of pollinators in culture room (Fig. 2b). Here, we need to

confirm whether and when the C. pumila flowers are self-

fertile through a series of experiments. We first dissected

longitudinally the flowers just close to anthesis to examine

whether the sexual organs are mature (Fig. 3a, b). Within the

closed corolla tube, the style is held in position pressed

against the upper inner surface of the tube with the bila-

mellar stigma curved downward at the tip (Fig. 3b, c). The

filaments of two stamens strongly geniculate at the mid-

point and lift the two face-to-face cohered anthers above the

stigma and pressing against the style (Fig. 3b, c). There is a

great amount of pollen released from anthers and fallen on

the lower inner surface of the corolla tube (Fig. 3b, c). The

results of in vitro pollen germination experiments further

showed that nearly 100 % of pollen grains started to ger-

minate after incubating on the culture medium at 28 �C for

2–3 h, and pollen tubes continued to elongate after 8 h of

incubation, indicative of the vitality of pollen (Fig. 3d).

Immersing stigmas in peroxide solution has been used

to measure the stigma receptiveness (Bredemeijer 1982).

In this experiment, the pistils were dissected from 24

flowers just close to anthesis and immersed into a 3 %

H2O2 solution for several minutes. Many oxygen bubbles

were formed and released from the stigma due to the pre-

sence of the peroxidase enzyme (Fig. 3e), indicating the

receptivity of the stigma.

To investigate the growth of pollen tubes in situ, flowers

close to anthesis were either directly collected or bagged in

Cellophane for further 2 days. The pistils were dissected

and stained with aniline blue. Under the fluorescent

microscope, a great number of pollen grains were found to

adhere to the stigmas of the flowers just close to anthesis

and the pollen tubes began to germinate (data not shown).

The pollen tubes were readily visualized on the stigmas of

the bagged flowers (Fig. 3f).

We further tested the self-fertilization of C. pumila

flowers by bagging experiment. The seed setting percent-

age was counted 2 weeks later. Of 60 flowers analyzed, 58

yielded fertile capsules (96.7 % of seed setting percentage;

see Supplementary Table 1). To examine whether C. pu-

mila could be cross-pollinated when flowers open, we

artificially emasculated seven immature flowers of about

1.5 cm in length (mature flowers are 3–4 cm in length),

and then artificially pollinated six flowers 48–72 h later

with one flower served as a negative control. About

2 weeks later, six hand-pollinated flowers all produced

Fig. 3 C. pumila flowers are self-pollinated. a–c A flower just close

to anthesis was dissected longitudinally to show its mature stamens.

Bars, 0.5 cm. d The pollen tubes germinated on culture medium.

Bar, 100 lm. e Activity analysis of the stigma of a flower just before

anthesis. Bar, 1 cm. f Germinated pollen tubes on the stigma of a

bagged flower. Bar, 200 lm

Plant Cell Tiss Organ Cult

123

fertile capsules, whereas the artificially emasculated but

non-pollinated flower failed to fruit (Supplementary

Fig. 1), indicating that C. pumila flowers have a potential

of cross-pollination, facilitating the genetic cross.

Karyotype and genome size analyses of C. pumila

The chromosome number and size of C. pumila were

evaluated by observing root tip cells at the mitotic meta

phase under a microscope (Fig. 4a, b). The result showed that

C. pumila has eight chromosomes of 3.5–6.0 lm in length

(Fig. 4c, d). The karyotype is formulated as 2n = 8 =

6 m ? 2 sm (2 sat) with three pairs of m-chromosomes and

one pair of sm-chromosomes (Fig. 4d; Supplementary

Table 2). The first pair of chromosomes has a secondary

constriction in the interstitial region of both the long and

short arms (Fig. 4d, No. 1 and 2), whereas the second one has

a secondary constriction only in the long arm (Fig. 4d, No. 3

and 4). Two sm-chromosomes have a satellite in the terminal

region of the short arm (Fig. 4d, No. 3 and 4).

We further measured the absolute nuclear DNA amount

(genome size) of C. pumila using flow cytometry. Three

different C. pumila plants were analyzed with rice (Oryza

sativa L. var. Nipponbare) served as an internal reference

standard. According to Burr (2002), the 2C DNA amount of

rice is 0.9 pg. In this experiment, the mean ratio of G1 peaks

(C. pumila : rice) was 1.8 (Supplementary Fig. 2). Therefore,

the 2C DNA amount of C. pumila was estimated to be 1.6 pg.

According to the formula 1 pg DNA = 0.978 9 109 bp

(Dolezel et al. 2007), the haploid genome size of C. pumila

was about 798.7 Mb (Supplementary Table 3).

Shoot induction and antibiotic sensitivity experiments

BA and NAA at different concentrations were added into MS

medium to assess their effects on shoot induction rate. Leaf

explants thickened after about 1 week of culture. About

2 weeks later, adventitious shoots began to appear on the

wound edges of the explants. While MS medium lacking

growth regulators led to a relatively low shoot induction rate

(67.8 ± 10.0 %), the addition of appropriate concentration

of BA and NAA enhanced it. The highest shoot induction rate

(97.6 ± 4.1 %) was obtained when 0.5 mg l-1 BA and

0.1 mg l-1 NAA were applied (Table 1). Under this condi-

tion, abundant adventitious shoots could be induced from

explants within 4 weeks with no appearance of undesirable

roots (Table 1; Supplementary Fig. 3). The result also

showed that relatively lower BA and higher NAA led to high

root induction rate, and it reached up to 95.1 ± 4.3 % when

1.0 mg l-1 NAA were applied (Table 1; Supplementary

Fig. 3). Nevertheless, root induction in C. pumila requires

neither BA nor NAA, and the adventitious shoots could

naturally generate roots after transferring to fresh MS med-

ium without any growth regulator (data not shown). There-

fore, 0.5 mg l-1 BA and 0.1 mg l-1 NAA were applied in

following shoot induction experiments, and MS medium

without any growth regulator was used for root induction.

It is reported that abundant adventitious shoots could be

induced from two Gesneriaceae plants at 25 �C (Tang et al.

2007a, b). Here, to evaluate whether different temperatures

affect the shoot induction rate, fresh leaf explants were

cultured on MS medium supplied with 0.5 mg l-1 BA and

0.1 mg l-1 NAA at different temperatures. The highest

shoot induction rate (97.8 ± 11.6 %) was obtained at

26 �C. Both higher and lower temperatures reduced the

shoot induction rate, and it dropped to 55.5 % at 22 �C

(Supplementary Table 4). However, Fisher’s LSD test

(P \ 0.05) showed that the shoot induction rates obtained

at 24, 26 and 28 �C were not significantly different from

each other, indicating that C. pumila can adapt to a rela-

tively wide temperature range. Nevertheless, for unifor-

mity, an intermediate temperature, i.e. 26 �C was applied

in following experiments.

Hygromycin at different concentrations was added into

MS medium containing 0.5 mg l-1 BA and 0.1 mg l-1

NAA to determine whether it is effective for selecting C.

pumila transformants. The induction of adventitious shoots

was severely affected by hygromycin at the concentration

of 10 mg l-1 or higher. When its concentration reached

to 20 mg l-1, all explants became necrotic with no shoot

Fig. 4 The karyotype analyses of C. pumila. a Resting nucleus.

Bar, 5 lm. b Mitotic prophase chromosomes. Bar, 5 lm. c The eight

chromosomes at the mitosis metaphase. Bar, 5 lm. d The karyotype

of C. pumila. Bar, 5 lm

Plant Cell Tiss Organ Cult

123

induced (Supplementary Table 5). Therefore, 20 mg l-1

hygromycin was used in following transformation

experiments.

Factors affecting hygromycin-resistant shoot induction

frequency in transformation experiments

Several factors affect Agrobacterium inoculation and shoot

induction, including Agrobacterium cell density, infection

time, pre-culture time and co-culture duration (Mondal et al.

2001; Kim et al. 2004; Barik et al. 2005; Crane et al. 2006; Du

and Pijut 2009; Jian et al. 2009). In this study, co-culture time

was first examined by infecting fresh leaf explants with

Agrobacterium LBA4404 of OD600 = 0.6 for 20 min and

culturing on the co-culture medium for 1–5 days. The

highest shoot induction rate (42.2 ± 5.1 %) was achieved

after 2-days of co-culture, and it declined with shortened or

prolonged co-culture (Fig. 5a). 1- and 5-days of co-culture

resulted in the lowest shoot induction rate. Hence, co-culture

for 2 days was applied to next transformation experiments.

To evaluate whether pre-culture could enhance the induc-

tion frequency, newly prepared leaf explants were pre-cul-

tured on MS medium for 0–4 days, inoculated with

Agrobacterium of OD600 = 0.6 for 20 min, and co-cultured

for 2 days. While fresh explants gave rise to the lowest shoot

induction rate (41.7 ± 3.6 %), pre-culture significantly

enhanced it (Fig. 5b). 1-day of pre-culture witnessed the

highest shoot induction rate (95.8 ± 7.2 %), whereas exten-

ded duration led to slightly lower rate. Even though Fisher’s

LSD test (P \ 0.05) showed that 1, 2, 3 or 4 days of pre-

culture resulted in no significant difference in the shoot

induction rate, a pre-culture of 1 day was applied in following

assays due to timesaving and high shoot induction rate.

To analyze whether the OD600 value of Agrobacterium

influences the induction frequency, 1-day pre-cultured leaf

explants were inoculated with Agrobacterium LBA4404 of

different OD600 values for 20 min and co-cultured for

2 days. The highest regeneration frequency (90.0 ± 3.2 %)

was achieved at the late-log phase (corresponding to

OD600 = 0.6), whereas both lower and higher OD600 val-

ues significantly reduced it (Fig. 5c). When OD600 reached

up to 1.0, the lowest regeneration efficiency (52.6 ±

2.8 %) was obtained because of the uncontrollable over-

growth of Agrobacterium.

To verify if Agrobacterium inoculation time affects the

shoot induction rate, 1-day pre-cultured leaf explants were

immersed in Agrobacterium LBA4404 solutions

(OD600 = 0.6) for 10–50 min. 20 min of inoculation was

found to achieve the highest shoot induction frequency

(92.5 ± 5.0 %) (Fig. 5d). It markedly decreased with

increased inoculation time, dropping to 30.0 ± 1.0 %

when the infection time was 50 min because of the over-

growth of Agrobacterium. Although the difference between

20 and 30 min of inoculation was not significant, a rela-

tively short infection period is probably more beneficial for

the viability of explants. Therefore, 20 min of inoculation

was used in following experiments.

GUS activity assays

57 Hyg-resistant plantlets from different leaf explants were

obtained using the optimal conditions, i.e. inoculating

Fig. 5 Effects of different

factors on the shoot induction

rate of C. pumila. Effects of co-

culture time (a), pre-culture

time (b), OD600 value (c), and

infection time (d) on the shoot

induction rate were determined.

Means of induction efficiencies

were compared using a Fisher’s

LSD test (P \ 0.05) and column

bars labeled with the same

letters are not significantly

different. Data, presented as

mean ± SD, were calculated

from three independent

experiments with about 40 leaf

explants each

Plant Cell Tiss Organ Cult

123

1-day pre-cultured explants with Agrobacterium LBA4404

(harboring pCAMBIA1301) for 20 min, culturing on

the co-culture medium for 2 days and selecting on the

selection medium containing 20 mg l-1 hygromycin.

Hygromycin-resistant shoots appeared on the wound edges

of explants after about 4 weeks of induction (Fig. 6a).

After further 4 weeks, the shoots of about 0.5 cm in length

were excised and transferred onto fresh MS medium

Fig. 6 Transgenic C. pumila plants and GUS activity assays.

a Adventitious shoots appeared on the wound edges of leaf explants.

The photo was taken 4 weeks after Agrobacterium inoculation. Bar,

1 cm. b Hygromycin-resistant shoots were transferred onto fresh MS

medium to promote rooting and shoot elongation. Bar, 1 cm. c Shoots

with obvious roots after maintained on MS medium for about 1 week.

Bar, 1 cm. d A transgenic plant (photographed 1 month after

transplantation). Bar, 1 cm. e PCR identification of transgenic plants.

M, 2 kb DNA marker; N, negative control; P, plasmid DNA; 1–7,

seven independent transgenic plants. f GUS staining of wild-type

(left) and transgenic (right) leaves. Bar, 1 mm. g GUS staining of

wild-type (left) and T0 transgenic (right) stems. Bar, 1 mm. h GUS

staining of wild-type (lane 1 and 2) and T1 transgenic (lane 3–6)

leaves

Plant Cell Tiss Organ Cult

123

(without any growth regulator) to promote rhizogenesis and

shoot elongation (Fig. 6b). About 1 week later, the roots

could be readily observed (Fig. 6c). Plantlets of 1–2 cm in

length with well-developed roots were transplanted into

pots. Transgenic plants grew well in culture room with

purple spots appearing slowly on the old leaves (Fig. 6d).

Furthermore, the transgenic plants overexpressing the GUS

gene were morphologically normal comparing with wild-

type ones (Supplementary Fig. 4).

Polymerase chain reaction was carried out to investigate

whether the GUS gene had been introduced into C. pumila.

As a result, specific gene products of expected size

(1,081 bp) were amplified from seven independent trans-

genic plants with no fragment amplified from wild-type

ones (Fig. 6e). GUS activity assays were conducted to

investigate the GUS expression. Both leaves and stems of

transgenic plants were analyzed with wild-type ones served

as negative controls to eliminate the possibility of endog-

enous GUS expression. The results showed that strong and

uniform GUS signal was observed in both leaves and stems

of transgenic plants with no expression signal in wild-type

ones (Fig. 6f, g).

Progenies of line 2 (Fig. 6e) were further analyzed to

validate the GUS gene inheritance by culturing on MS

selection medium containing 25 mg l-1 hygromycin. Of 33

progenies analyzed, 24 could germinate and develop nor-

mally, while the remainder slowly became withered after

germination, conforming to a Mendelian segregation ratio

(3:1) for monogenic inheritance. Subsequent GUS activity

assays showed that these hygromycin-resistant seedlings

could generate GUS signal with no expression signal in

wide-type ones. The above results clearly indicated that the

GUS gene had been introduced into the host and obtained a

stable and uniform expression.

Validation of the transformation system by transferring

the GsNST1B gene

GsNST1B controlling the secondary wall biosynthesis in G.

soja (Dong et al. 2013) was introduced into C. pumila to

validate the availability of this transformation system in

gene function investigation. Positive transgenic plantlets

were confirmed by PCR followed by DNA sequencing to

avoid the possible amplification of endogenous NST-like

genes. The results showed that of 199 hygromycin-resistant

plantlets induced from 55 different leaf explants (for each

explant, 3–4 plants were analyzed by PCR), 23 belonging to

6 different transgenic lines were confirmed to be positive. In

contrast to wild-type plants that had normally developed

leaves (Fig. 7a), transgenic plants showed upward curling

leaves (Fig. 7b–d). Sections of the transgenic and wild-type

leaves were stained with toluidine blue to understand the

cellular basis of upward curling leaves. In wild-type leaves,

the secondarily thickened cells were only found in the veins

and xylem strands (Fig. 7e). In contrast, the parenchyma

mesophyll cells, in addition to the veins and xylem strand

cells were heavily secondarily thickened in the transgenic

leaves (Fig. 7f). RT-PCR was further carried out to check

the expression of the GsNST1B gene in transgenic plants. As

shown in Fig. 7g, the GsNST1B gene was strongly expressed

in three independent transgenic plants with no signal in

wild-type ones. In addition, GsNST1B overexpressors gen-

erated undeveloped fruits (data not shown), similar to our

previous report (Dong et al. 2013).

Discussion

Characterization of C. pumila as an ideal model plant

for evo-devo studies

The haploid chromosome number n = 4 of C. pumila was

previously reported by Ratter (1963). We here document its

karyotype 2n = 8 = 6 m ? 2sm (2sat), the lowest number

of chromosomes reported in Gesneriaceae to date (Skog

1984; Li and Wang 2004; Weber 2004; this study). The

species in Gesneriaceae are usually polyploidy with

perennial habit (Skog 1984; Li and Wang 2004; Weber

2004). The genus Chirita sensu stricto is one of the rarely

occurred diploid taxa with annual habit in Gesneriaceae

(the traditional polyphyletic Chirita was split into four

monophyletic groups including the perennial Primulina

and Liebigia and the annual Chirita sensu stricto and

Microchirita; see Wang et al. 2011). The close relatives of

C. pumila in Chirita usually have diploid chromosome

number of 2n = 18 (Li and Wang 2004). Given that the

basic chromosome number is x = 9 or 8 in Gesneriaceae,

the low chromosome number of C. pumila was assumed to

be achieved through successive unequal translocation from

a complement of chromosomes with x = 9, probably cor-

related with its short-lived habit (Ratter 1963; Skog 1984;

Li and Wang 2004). Researches addressing chromosome

evolution between C. pumila and its relatives would reveal

how the rearranged chromosomes contribute to plant habit

shifts, reproductive isolation and speciation.

In addition, we found a special phenomenon that C.

pumila flowers perform self-fertilization before anthesis,

i.e. cleistogamy, an extreme form of self-fertilization first

reported in Gesneriaceae. Selfing has commonly been

viewed as an ‘‘evolutionary dead end’’ because it usually

leads to inbreeding depression by accumulating recessive

deleterious alleles (Stebbins 1957; Barrett 2002; Boggs

et al. 2009). However, when environments become fluc-

tuant and unpredictable with scarcity or inconsistency of

pollinators or population bottlenecks, selfing rates would

increase with inbreeding depression gradually overcome

Plant Cell Tiss Organ Cult

123

because deleterious alleles may be purged over successive

generations of selfing (Barrett 2002; Boggs et al. 2009;

Albert et al. 2011). The self-fertilization will finally

become established owing to adaptive advantages of self-

pollination in providing reproductive assurance when out-

crossing fails (Darwin 1876). Morphologically, the con-

version from outcrossing to cleistogamy involves the shifts

from dichogamy to homogamy and from herkogamy to

plesiogamy, and precocious maturation of sexual organs

(Campbell et al. 1983). In Gesneriaceae, almost all species

have hermaphroditic flowers containing both female and

male sexual organs which usually spatially separate in a

flower, i.e. herkogamy for cross-pollination. Sexually

mature pollen and stigmas are presented as flowers bloom

to attract animal pollinators (Wang et al. 2010, 2011). In

the evolutionary transition from cross-fertilization to

cleistogamy, a series of floral morphological and physio-

logical modifications have occurred in C. pumila, including

the same position of the anthers and stigmas just below the

upper inner surface of the corolla tube, and the precocious

and simultaneous maturation of pollen and stigmas before

anthesis. In addition, the flowers of C. pumila open with

anthers included and stigma exserted. Given hand-

pollinated flowers producing fertile fruits, C. pumila

should have the possibility of cross-fertilization, a mixed

mating system envisaged as a ‘‘bet-hedging strategy’’ for

fluctuating and unpredictable environments (Berg and

Redbo-Torstensson 1998; Culley and Klooster 2007). The

typical cleistogamous flowers with potential cross-polli-

nation make C. pumila an ideal candidate model to

understand the ecological success of natural selection and

genetic mechanisms for the mating systems of cleistogamy

versus chasmogamy.

The ABCE model is a widely used framework to

understand the floral development and evolution in angio-

sperms (Soltis et al. 2007; Litt and Kramer 2010). How-

ever, some components of the ABCE model, such as

A-function floral identity genes, are so far limited to

Arabidopsis and its close relatives and their functions have

not yet been testified in other lineages of angiosperms (e.g.

Antirrhinum; Litt 2007; Bowman et al. 2012). Additional

function analyses in emerging evo-devo model organisms

are therefore critically important to finally elucidate whe-

ther the BC model lacking the A-function is general in

eudicots (Litt 2007; Soltis et al. 2007; Causier et al. 2010;

Bowman et al. 2012). In addition, the origin of zygomor-

phic flowers is suggested to be one key innovation asso-

ciated with the explosive radiation of angiosperms (Dilcher

2000; Cubas 2004; Busch and Zachgo 2009). Increasing

evidence indicates that CYC-like TCP genes play a crucial

role in the origin and evolution of floral zygomorphy in

angiosperms (Cubas 2004; Busch and Zachgo 2009;

Fig. 7 Analysis of transgenic plants overexpressing GsNST1B. a–dOne wild-type plant (a) and three independent transgenic plants (b–d)with upward curling leaves. Bars, 1 cm. e Toluidine blue staining

results showing normal secondary wall thickening of wild-type

plants in the veins and vascular bundles. Bar, 10 lm. f The ectopic

secondary wall thickening of transgenic plants in parenchyma

mesophyll cells. Bar, 10 lm. g RT-PCR analysis of GsNST1B in

three independent transgenic plants with wild-type plant served as a

negative control. CpACTIN was amplified as an internal reference

Plant Cell Tiss Organ Cult

123

Preston and Hileman 2009; Specht and Bartlett 2009). A

recent report suggests that the repeated origins of floral

zygomorphy are related to the independent gains of similar

positive autoregulatory elements in CYC-like TCP genes in

different lineages of angiosperms (Yang et al. 2012).

However, it is still a challenge to decipher how these genes’

activities are controlled by upstream factors, including the

dorsal identity function, the functional domain expansion to

lateral or ventral floral regions, the loss-of-function and so

on (Song et al. 2009; Martın-Trillo and Cubas 2010; Yang

et al. 2012; Hileman 2014). C. pumila is apparently an ideal

model species to address these questions because of its

phylogenetic representativeness, annual habit, short life

cycle, and self-fertility and diploid, the widely accepted

selection criteria for evo-devo model organisms (Irish and

Benfey 2004; Jenner 2006; Jenner and Wills 2007; Sommer

2009; Ankeny and Leonelli 2011). Its diploid with low

chromosome number would facilitate the identification of

recessive traits and avoid the complication of gene dosages,

and its typical cleistogamous flowers with potential cross-

pollination enable C. pumila to be maintained with homo-

zygous lines straightforward and capable of generating

genetic crosses. Therefore, we here select C. pumila as a

target to develop the genetic transformation system.

Efficient Agrobacterium-mediated transformation

and regeneration of C. pumila

An efficient Agrobacterium-mediated transformation and

regeneration system is developed in C. pumila in this study,

which depends on a powerful shoot induction ability of

leaf explants on MS medium containing 0.5 mg l-1 BA

and 0.1 mg l-1 NAA (Table 1). Moreover, similar to

Perilla frutescens (Kim et al. 2004), the adventitious shoots

of C. pumila directly form on the wound edges of leaf

explants without an evident callus phase. In addition, C.

pumila has a powerful rooting ability because no growth

regulator is required and roots can be readily observed after

1 week of culture on fresh MS medium. Therefore, C.

pumila has a powerful regeneration ability using leaf

explants, a prerequisite for genetic transformation.

High efficient gene transfer and powerful regeneration

ability of explants after Agrobacterium inoculation are

crucial to plant transformation (He et al. 2010). Many

factors may affect the transformation frequency and

thereafter the shoot induction rate to varying degrees. As a

pivotal step in transformation process, co-culture of the

inoculated explants with Agrobacterium allows T-DNA

transfer from plasmid into plant cells. In general, co-culture

for 2–3 days reaches the highest transformation frequency

(Kim et al. 2004; Jian et al. 2009). However, co-culture

duration can be prolonged to 4–5 days for some species

(Mondal et al. 2001; Barik et al. 2005). Here, 2 days of

co-culture reaches the highest shoot induction rate. While

1 day of co-culture is insufficient for Agrobacterium

infection and T-DNA transfer, extended duration may

cause the damage of explants owing to the uncontrollable

overgrowth of Agrobacterium. Accordingly, the shoot

induction frequency is reduced in both cases.

It is reported that pre-culture of explants prior to

Agrobacterium inoculation can significantly enhance the

transformation frequency in Cajanus cajan (Lawrence and

Koundal 2000), Lathyrus sativus (Barik et al. 2005), Lotus

corniculatus (Jian et al. 2009) and Fagopyrum esculentum

(Chen et al. 2008). However, pre-culture drastically

declines the transformation competence of Citrus paradise

(Costa et al. 2002) and Perilla frutescens (Kim et al. 2004).

In this study, pre-culture enhances drastically the shoot

induction rate, indicating a positive effect of pre-culture on

C. pumila transformation probably due to the improved

viability of explants. However, extended pre-culture redu-

ces slightly the shoot induction rate probably by dimin-

ishing the susceptibility of explants to Agrobacterium,

indicating that only appropriate pre-culture duration is

benefit for the transformation and regeneration of

C. pumila.

Agrobacterium cell density and inoculation time can

also affect transformation efficiency (Du and Pijut 2009;

Jian et al. 2009). While low Agrobacterium cell concen-

tration and short infection time may result in insufficient

attachment of Agrobacterium to explants and reduce the

transformation frequency, increased Agrobacterium cells

and prolonged inoculation time would damage explants

and decrease the regeneration frequency. Here, the highest

shoot induction frequency is obtained when fresh leaf

explants are infected with Agrobacterium cells of

OD600 = 0.6 for 20 min. Both high and low Agrobacte-

rium cell density, as well as shortened or prolonged inoc-

ulation time reduce the shoot induction rate, indicating that

appropriate Agrobacterium cell density and inoculation

time are important for successful transformation.

As outlined above, a high shoot induction frequency is

achieved in C. pumila by infecting 1-day pre-cultured

explants with Agrobacterium of OD600 = 0.6 for 20 min

followed by 2-days of co-culture. Further PCR and GUS

activity assays of T0 and T1 plants indicate that the GUS

gene has been successfully introduced into the host, evident

by its stable and uniform expression. However, further

transformation experiment using a gene with known

function and obvious phenotypic effects is required to

validate the applicability of a transformation system in

gene function investigation (Jian et al. 2009). Here, the

GsNST1B gene implicated in secondary wall biosynthesis

in G. soja is selected due to its obvious phenotype during

early vegetative growth stages in Arabidopsis overexpres-

sors (Dong et al. 2013). Similar to Dong et al. (2013),

Plant Cell Tiss Organ Cult

123

GsNST1B overexpression in C. pumila generates a desired

phenotype characteristic of upward curling rosette leaves

that is attributed to the ectopic thickening of secondary

walls, indicating the applicability of this transformation

system in gene function investigation.

A reliable and efficient transformation system is crucial to

comparative functional studies in evo-devo that aims at

exploring the evolutionary mechanisms underlying morpho-

logical changes. However, developing a transgenic system is

usually time-consuming and laborious because it requires

several generations of subculture and alterations of medium.

In Lotus japonicas and Medicago truncatula, for example,

about 4 months are needed to produce transgenic plants

(Stiller et al. 1997; Crane et al. 2006). In Lotus corniculatus,

the superroot- derived transformation protocol is complicated

with at least five changes of medium (Jian et al. 2009).

In Triticum turgidum, obtaining transgenic lines requires 2–3

rounds of selection (He et al. 2010). As a classical model plant,

Antirrhinum has been proved to be successful in Agrobacte-

rium-mediated genetic transformation accompanied by

repeatedly improved transformation protocol (Cui et al. 2003,

2004; Manchado-Rojo et al. 2012). Nevertheless, it still leaves

something to be desired that might restrict its wide applica-

tion. In this study, using leaf disks as explants, the transfor-

mation process (from Agrobacterium inoculation to PCR

identification) takes about 3 months with a high efficiency

(Fig. 8). In addition, the entire transformation process is

simple because no specific rooting media is required and the

shoot induction and selection are achieved in one step. Taken

together, the C. pumila transformation system has the features

of simplicity, rapidity and high-efficiency.

Since its inception at the end of last century, evo-devo

has passed from an initial stage to a rapid developing

discipline, evident by emerging model organisms in both

animals and plants. With completion of genome sequenc-

ing project, establishment of a mutant library and further

optimization of the transformation system, C. pumila could

become a fascinating model plant for a wide range of evo-

devo studies, especially in the field of floral symmetry,

floral organ identity, chromosome evolution and mating

system evolution.

Acknowledgments We thank James F. Smith for his constructive

comments and language improvements on this article. This work was

supported by the National Natural Science Foundation of China

(30990240 and 31170198).

References

Albert LP, Campbell LG, Whitney KD (2011) Beyond simple

reproductive assurance: cleistogamy allows adaptive plastic

responses to pollen limitation. Int J Plant Sci 172(7):862–869

Ankeny RA, Leonelli S (2011) What’s so special about model

organisms? Stud Hist Philos Sci 42:313–323

Baguna J, Garcia-Fernandez J (2003) Evo-devo: the long and winding

road. Int J Dev Biol 47:705–713

Barik DP, Mohapatra U, Chand PK (2005) Transgenic grasspea

(Lathyrus sativus L.): factors influencing Agrobacterium-medi-

ated transformation and regeneration. Plant Cell Rep 24:523–531

Barrett SCH (2002) The evolution of plant sexual diversity. Nat Rev

Genet 3:274–284

Berg H, Redbo-Torstensson P (1998) Cleistogamy as a bet-hedging

strategy in Oxalis acetosella, a perennial herb. J Ecol 86:491–500

Bliss BJ, Wanke S, Barakat A et al (2013) Characterization of the

basal angiosperm Aristolochia fimbriata: a potential experimen-

tal system for genetic studies. BMC Plant Biol 13:13

Boggs NA, Nasrallah JB, Nasrallah ME (2009) Independent S-locus

mutations caused self-fertility in Arabidopsis thaliana. PLoS

Genet 5(3):e1000426. doi:10.1371/journal.pgen.1000426

Bowman JL, Smyth DR, Meyerowitz EM (2012) The ABC model of flower

development: then and now. Development 139(22):4095–4098

Bredemeijer GMM (1982) Mechanism of peroxidase isoenzyme

induction in pollinated Nicotiana alata styles. Theor Appl Genet

62:305–309

Breuker CJ, Debat V, Klingenberg CP (2006) Functional evo-devo.

Trends Ecol Evol 21:488–492

Burr B (2002) Mapping and sequencing the rice genome. Plant Cell

14:521–523

Busch A, Zachgo S (2009) Flower symmetry evolution: towards

understanding the abominable mystery of angiosperm radiation.

BioEssays 31:1181–1190

Campbell CS, Quinn JA, Cheplick GP, Bell TJ (1983) Cleistogamy in

grasses. Annu Rev Ecol Syst 14:411–441

Carroll SB (2008) Evo-devo and an expanding evolutionary synthesis:

a genetic theory of morphological evolution. Cell 134:25–36

Causier B, Schwarz-Sommer Z, Davies B (2010) Floral organ

identity: 20 years of ABCs. Semin Cell Dev Biol 21:73–79

Chen L-H, Zhang B, Xu Z-Q (2008) Salt tolerance conferred by

overexpression of Arabidopsis vacuolar Na?/H? antiporter gene

AtNHX1 in common buckwheat (Fagopyrum esculentum).

Transgenic Res 17:121–132

Fig. 8 A flowchart for Agrobacterium-mediated transformation of

C. pumila

Plant Cell Tiss Organ Cult

123

Costa MGC, Otoni WC, Moore GA (2002) An evaluation of factors

affecting the efficiency of Agrobacterium-mediated transformation

of Citrus paradise (Macf.) and production of transgenic plants

containing carotenoid biosynthetic genes. Plant Cell Rep 21:365–373

Crane C, Wright E, Dixon RA, Wang Z-Y (2006) Transgenic

Medicago truncatula plants obtained from Agrobacterium tum-

efaciens-transformed roots and Agrobacterium rhizogenes-trans-

formed hairy roots. Planta 223:1344–1354

Cubas P (2004) Floral zygomorphy, the recurring evolution of a

successful trait. BioEssays 26:1175–1184

Cui M-L, Handa T, Ezura H (2003) An improved protocol for

Agrobacterium-mediated transformation of Antirrhinum majus

L. Mol Genet Genomics 270:296–302

Cui M-L, Ezura H, Nishimura S, Kamada H, Handa T (2004) A rapid

Agrobacterium-mediated transformation of Antirrhinum majus

L. by using direct shoot regeneration from hypocotyl explants.

Plant Sci 166:873–879

Culley TM, Klooster MR (2007) The cleistogamous breeding system:

a review of its frequency, evolution, and ecology in angiosperms.

Bot Rev 73:1–30

Darwin CR (1876) The effects of cross and self-fertilization in the

vegetable kingdom, Ch 9. John Murray, London

de Bruijn S, Angenent GC, Kaufmann K (2012) Plant ‘evo-devo’ goes

genomic: from candidate genes to regulatory networks. Trends

Plant Sci 17:441–447

Dilcher D (2000) Toward a new synthesis: major evolutionary trends

in the angiosperm fossil record. Proc Natl Acad Sci USA 97:

7030–7036

Dolezel J, Greilhuber J, Suda J (2007) Estimation of nuclear DNA

content in plants using flow cytometry. Nat Protoc 2:2233–2244

Dong Y, Wang B-H, Wang Y-Z (2013) Functional characterization of

the orthologs of AtNST1/2 in Glycine soja (Fabaceae) and the

evolutionary implications. J Syst Evol 51(6):693–703

Dong Y, Yang X, Liu J, Wang B-H, Liu B-L, Wang Y-Z (2014) Pod

shattering resistance associated with domestication is mediated

by a NAC gene in soybean. Nat Commun 5:3352

Donoghue MJ, Ree RH, Baum DA (1998) Phylogeny and the

evolution of flower symmetry in the Asteridae. Trends Plant Sci

3:311–317

Du N, Pijut PM (2009) Agrobacterium-mediated transformation of

Fraxinus pennsylvanica hypocotyls and plant regeneration. Plant

Cell Rep 28:915–923

Du Z-Y, Wang Y-Z (2008) Significance of RT-PCR expression

patterns of CYC-like genes in Oreocharis benthamii (Gesneri-

aceae). J Syst Evol 46:23–31

Endress PK (1998) Antirrhinum and Asteridae—evolutionary changes

of floral symmetry. Symp Soc Exp Biol 53:133–140

Galbraith DW, Harkins KR, Maddox JM, Ayres NM, Sharma DP,

Firoozabady E (1983) Rapid flow cytometric analysis of the cell

cycle in intact plant tissues. Science 220:1049–1051

Gao Q, Tao J-H, Yan D, Wang Y-Z, Li Z-Y (2008) Expression

differentiation of CYC-like floral symmetry genes correlated

with their protein sequence divergence in Chirita heterotricha

(Gesneriaceae). Dev Genes Evol 218:341–351

He Y, Jones HD, Chen S, Chen XM, Wang DW, Li KX, Wang DS,

Xia LQ (2010) Agrobacterium-mediated transformation of

durum wheat (Triticum turgidum L. var. durum cv Stewart)

with improved efficiency. J Exp Bot 61:1567–1581

Hileman LC (2014) Bilateral flower symmetry—how, when and why?

Curr Opin Plant Biol 17:146–152

Irish VF, Benfey PN (2004) Beyond Arabidopsis. Translational

biology meets evolutionary developmental biology. Plant Phys-

iol 135:611–614

Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: beta-

glucuronidase as a sensitive and versatile gene fusion marker in

higher plants. EMBO J 6:3901–3907

Jeffery WR (2008) Emerging model systems in evo-devo: cavefish

and microevolution of development. Evol Dev 10:265–272

Jenner RA (2006) Unburdening evo-devo: ancestral attractions, model

organisms, and basal baloney. Dev Genes Evol 216:385–394

Jenner RA, Wills MA (2007) The choice of model organisms in evo-

devo. Nat Rev Genet 8(4):311–319

Jian B, Hou W, Wu C, Liu B, Liu W, Song S, Bi Y, Han T (2009)

Agrobacterium rhizogenes-mediated transformation of Super-

root-derived Lotus corniculatus plants: a valuable tool for

functional genomics. BMC Plant Biol 9:78

Jiang LX, Yang SL, Xie LF, San Puah C, Zhang XQ, Yang WC,

Sundaresan V, Ye D (2005) VANGUARD1 encodes a pectin

methyl esterase that enhances pollen tube growth in the

Arabidopsis style and transmitting tract. Plant Cell 17:584–596

Kellogg EA (2004) Evolution of developmental traits. Curr Opin

Plant Biol 7:92–98

Kellogg EA (2006) Progress and challenges in studies of the evolution

of development. J Exp Bot 57:3505–3516

Kim K-H, Lee Y-H, Kim D, Park Y-H, Lee J-Y, Hwang Y-S, Kim

Y-H (2004) Agrobacterium-mediated genetic transformation of

Perilla frutescens. Plant Cell Rep 23:386–390

Kopp A (2009) Metamodels and phylogenetic replication: a system-

atic approach to the evolution of developmental pathways.

Evolution 63:2771–2789

Kramer EM (2007) Understanding the genetic basis of floral diversity.

Bioscience 57:479–487

Kushikawa S, Hoshino Y, Mii M (2001) Agrobacterium-mediated trans-

formation of Saintpaulia ionantha Wendl. Plant Sci 161:953–960

Lawrence PK, Koundal KR (2000) Simple protocol for Agrobacte-

rium tumefaciens-mediated transformation of pigeon pea [Caj-

anus cajan (L.) Millsp.]. J Plant Biol 27:299–302

Li Z-Y, Wang Y-Z (2004) Plants of Gesneriaceae in China. Henan

Science and Technology Publishing House, Zhengzhou

Litt A (2007) An evaluation of A-function: evidence from the APET-

ALA1 and APETALA2 gene lineages. Int J Plant Sci 168:73–91

Litt A, Kramer EM (2010) The ABC model and the diversification of

floral organ identity. Semin Cell Dev Biol 21:129–137

Liu B-L, Pang H-B, Yang X, Wang Y-Z (2014) Functional and

evolutionary analyses of Primulina heterotricha CYC1C gene in

tobacco and Arabidopsis transformation systems. J Syst Evol

52(1):112–123

Manchado-Rojo M, Delgado-Benarroch L, Roca MJ, Weiss J, Egea-

Cortines M (2012) Quantitative levels of Deficiens and Globosa

during late petal development show a complex transcriptional

network topology of B function. Plant J 72:294–307

Mandoli DF, Olmstead R (2000) The importance of emerging model

systems in plant biology. J Plant Growth Regul 19:249–252

Martın-Trillo M, Cubas P (2010) TCP genes: a family snapshot ten

years later. Trends Plant Sci 15:31–39

Mercuri A, De Benedetti L, Burchi G, Schiva T (2000) Agrobacte-

rium-mediated transformation of African violet. Plant Cell

Tissue Organ Cult 60:39–46

Mondal TK, Bhattacharya A, Ahuja PS, Chand PK (2001) Transgenic

tea [Camellia sinensis (L.) O Kuntze cv. Kangrajat] plants

obtained by Agrobacterium-mediated transformation of somatic

embryos. Plant Cell Rep 20:712–720

Mori T, Karoiwa H, Higashiyama T, Kuroiwa T (2006) GENERA-

TIVE CELL SPECIFIC 1 is essential for angiosperm fertiliza-

tion. Nat Cell Biol 8:63–71

Muller GB (2007) Evo-devo: extending the evolutionary synthesis.

Nat Rev Genet 8:943–949

Murashige T, Skoog F (1962) A revised medium for rapid growth and

bio assays with tobacco tissue cultures. Physiol Plant 15:473

Pang H-B, Sun Q-W, He S-Z, Wang Y-Z (2010) Expression pattern of

CYC-like genes relating to a dorsalized actinomorphic flower in

Tengia (Gesneriaceae). J Syst Evol 48:309–317

Plant Cell Tiss Organ Cult

123

Preston JC, Hileman LC (2009) Developmental genetics of floral

symmetry evolution. Trends Plant Sci 14:147–154

Pruitt RE, Bowman JL, Grossniklaus U (2003) Plant genetics: a

decade of integration. Nat Genet Suppl 33:294–304

Ratter JA (1963) Some chromosome members in the Gesneriaceae.

Notes from the Royal Botanic Garden, Edinburgh

Skog LE (1984) A review of chromosome numbers in the Gesneri-

aceae. Selbyana 7:252–273

Soltis DE, Ma H, Frohlich MW et al (2007) The floral genome: an

evolutionary history of gene duplication and shifting patterns of

gene expression. Trends Plant Sci 12:358–367

Sommer RJ (2009) The future of evo-devo: model systems and

evolutionary theory. Nat Rev Genet 10:416–422

Song C-F, Lin Q-B, Liang R-H, Wang Y-Z (2009) Expressions of

ECE-CYC2 clade genes relating to abortion of both dorsal and

ventral stamens in Opithandra (Gesneriaceae). BMC Evol Biol

9:244

Specht CD, Bartlett ME (2009) Flower evolution: the origin and

subsequent diversification of the angiosperm flower. Annu Rev

Ecol Evol Syst 40:217–243

Stebbins GL (1957) Self-fertilization and population variability in the

higher plants. Am Nat 91:337–354

Stern DL (2000) Perspective: evolutionary developmental biology

and the problem of variation. Evolution 54:1079–1091

Stiller J, Martirani L, Tuppale S, Chian R-J, Chiurazzi M, Gresshoff

PM (1997) High frequency transformation and regeneration of

transgenic plants in the model legume Lotus japonicus. J Exp

Bot 48:1357–1365

Tang Z, Lin H, Shi L, Chen W (2007a) Rapid in vitro multiplication

of Chirita longgangensis W. T. Wang: an endemic and

endangered Gesneriaceae species in China. HortScience

42:638–641

Tang Z, Shi L, Chen W, Lin H (2007b) In vitro propagation of Chirita

heterotricha Merr. Propag Ornam Plants 7:43–48

Toth S, Kiss C, Scott P, Kovacs G, Sorvari S, Toldi O (2006)

Agrobacterium-mediated genetic transformation of the

desiccation tolerant resurrection plant Ramonda myconi (L.)

Rchb. Plant Cell Rep 25:442–449

Wang W, Pan K, Li Z, Weitzman AL, Skog LE (1998) Gesneriaceae.

Flora China 18:244–401

Wang Y-Z, Liang R-H, Wang B-H, Li J-M, Qiu Z-J, Li Z-Y, Weber A

(2010) Origin and phylogenetic relationships of the Old World

Gesneriaceae with actinomorphic flowers inferred from ITS and

trnL-trnF sequences. Taxon 59(4):1044–1052

Wang Y-Z, Mao R-B, Liu Y, Li J-M, Dong Y, Li Z-Y, Smith JF

(2011) Phylogenetic reconstruction of Chirita and allies (Ges-

neriaceae) with taxonomic treatments. J Syst Evol 49:50–64

Weber A (2004) Gesneriaceae. In: Kubitzki K, Kadereit JW (eds) The

families and genera of vascular plants. Volume VII, flowering

plants: Dicodyledons; Lamiales (except Acanthaceae including

Avicenniaceae). Springer, Berlin, pp 63–158

Wortley AH, Rudall PJ, Harris DJ, Scotland RW (2005) How much

data are needed to resolve a difficult phylogeny? Case study in

Lamiales. Syst Biol 54:697–709

Yang X, Cui H, Yuan Z-L, Wang Y-Z (2010) Significance of

consensus CYC-binding sites found in the promoters of both

ChCYC and ChRAD genes in Chirita heterotricha (Gesneria-

ceae). J Syst Evol 48:249–256

Yang X, Pang H-B, Liu B-L, Qiu Z-J, Gao Q, Wei L, Dong Y, Wang

Y-Z (2012) Evolution of double positive autoregulatory feed-

back loops in CYCLOIDEA2 clade genes is associated with the

origin of floral zygomorphy. Plant Cell 24:1834–1847

Zhang MZ, Ye D, Wang LL, Pang JL, Zhang YH, Zheng K, Bian

HW, Han N, Pan JW, Wang JH, Zhu MY (2008) Overexpression

of the cucumber LEAFY homolog CFL and hormone treatments

alter flower development in gloxinia (Sinningia speciosa). Plant

Mol Biol 67:419–427

Zhou X-R, Wang Y-Z, Smith JF, Chen RJ (2008) Altered expression

patterns of TCP and MYB genes relating to the floral develop-

mental transition from initial zygomorphy to actinomorphy in

Bournea (Gesneriaceae). New Phytol 178:532–543

Plant Cell Tiss Organ Cult

123